Cancer or tumors are brought on by abnormal cell growth and multiplication. The type of cancer is determined by where in the body aberrant cells first develop. Lung cancer is the malignancy that leads to the majority of cancer-related fatalities. Lung cancer is one of the majority frequent, the maximum mortality as well as death rates of any cancer in the globe. 2.1 million new cases in addition to 1.8 million deaths are already announced by Global Cancer Statistics in 2023. About one in five (18.4%) of lung cancer causes death. On the cancer side, medical imaging plays an important role in decreasing mortality. Treatment is a non-invasive as well as painless process through minimal aftereffect for patients. It can offer comprehensive anatomical data about the disease by creating a visual image of the cancer. Disparate previous invasive process for instance surgery and biopsy, which extracts and examines a tiny segment of the cancer tissue, imaging can provide a complete view as well as examination of the complete cancer region. In addition, therapeutic procedures requiring the analysis of tumors during treatment are preferred. In this study, the feature extraction, segmentation, and CNN methods are used to detect lung cancer using CT images. The Artificial Bee Colony (ABC) is employed in this case to segment the data. For feature extraction, features like Zernike and SIFT are used. Using principal component analysis (PCA), the condensed features (PCA) and the Convolutional Neural Network (CNN) classifier is an expert at identifying the typical tissue and the atypical tissue, and it is used to classify samples using Zernike and SIFT Features. The results show that the suggested technique is capable of accurately Classify identifying the typical tissue and the atypical tissue in Lung images.